Ah ok. Can I change this?
You could try testing with reduced distance and sensitivity in order to try and determine where the motion might be being detected. Donāt be afraid to try 0.15M distance and work your way higher.
I never really managed to get my sensor tuned in properly when I setup the device, just got back from holiday and decided Iād give calibration another go and make use of it. Surprised to see so much has changed since I originally set the device up! Thanks for all the hard work, glad itās easier to update the code now without needing to copy and paste my config to the top too. Now to just remember to check back now and again!
I have a feeling this will be a common trend since we (collectively) have not had the experience working with sensors that record any small movement. We wanted ābetter than PIRā and now that we have it, oh boyā¦ haha
Another common ask, from myself included, has been for āmore coverage.ā Hence me getting a 100Ā°x100Ā° version. But we might all be going in the wrong direction.
I put that sensor in a larger room of mine, and well, it catches āeverythingā.
Think I might try a different approach next and try and get a couple of the ānarrow FOVā version (Angle: 60Ā°Ć40Ā°) for a more targeted [sic] approach.
Yeah the coverage issue is the kicker for me really. Our living room is looking like weāll require at least 2 of these for full accurate coverage, even though itās only 5.5m x 5.5m (in an L shape). Iām getting detection stood up but as soon as I sit on the sofa it loses it sadly, I imagine due to the vertical detection range. For now Iāll probably just chuck it on the TV stand and detect if someone is on the sofa at the very least. I do wonder how good the Aqara ones are as a packaged solution.
Actually itād be an awesome comparison between bothā¦ I have 2 df robot that I have to setup, but wouldnāt mind to pay for the aqara if the performance is similar
He/She/They who designed L-shaped rooms did not consider us!
Iām working on the actual impact of anglesā¦ ā¦mine will detect at 90Ā° or -270Ā° facing front if you are <1M
Whatās the āstraight-line distanceā from sensor to sofa? And roughly what angle from sensor face?
I donāt have one. And I keep reading conflicting reports of ātakes 2-5 seconds to detection motion/presenceā. Maybe motion, maybe presenceā¦ ā¦hard to keep track!
I have the aqara as well as 2 x ācrlogicā sensors.
The only benefit of the aqara is that it is prepackaged. The ācrlogicā sensor beats it in every other way. The aqara does take about 5 seconds for me (compared side by side with a hue motion sensor) and it needs an occasional presence reset switch to be flipped else it gets stuff in detected mode.
I will at some point try to pickup the wider angle sensor, but other than that the ācrlogicā is fantastic.
Youāve done a huge amount of work on this with skills I can only dream of, and shared the output with the community. If I could nominate you for a contributions to home assistant award I would do so instantly.
Thank you.
Thanks for your reviewā¦ I wonāt spend my money in aqara that arenāt usually what I expect them to be! Just have to manage to find some good guide to calibrate my 3d printer or just order it online lol
Anyone else have issues with the binary sensor āBecame unavailableā after a period of time. It appears the device reboots, the red light lights up for a few seconds on the SEN0395. Anyone have any ideas? I have two of these that do the same thing, both based on Wemos D1 and otherwise running without issue. Distance at 3m and timeout at 10s, sensitivity 4.
Iāve seen a bunch of threads on the topic and at times experienced it myself. I do not believe that I have seen a consensus on a single cause/resolution.
So while Iāve found a recipe that works for me, no guarantees it works for you. Ever since using the following, Iāve not had any repeat issues across a dozen and a half ESPHome deployments.
Recipe
WIFI Settings
- a separate 2.4Ghz SSID the eternal debate, take it to Reddit
** only enable the basics on this SSID (no Fast roaming/transition etc.)
** DTIM 1
** power saving off
WIFI Coverage
- ensure all ESP devices are better than -70db RSSI
** yes can they work below that. Has it been as stable for me over the long-term? No. Will one person swear itās fine for them? Yes. lol
*** if someone can find the chart that I remember seeing, but cannot find again, where IDF uses a -70db trigger for roaming, let me know.
General Network Principle
- I prefer static IPās and PiHoleās while not relying upon mDNS.
ESPHome Settings
- static IP
- mDNS disabled
- wifi power_saving none
I realize that I just posted nine opinions that someone will loatheā¦ ā¦what have i done!
Thanks, I had already started down the path of static IP reservations for the devices after I posted the message, but within a few minutes one rebooted again just to laugh in my face
I will work through some of the other ideas. Iām using unifi APs along with OPNsense so I have all kinds of ways to break adjust the wireless. If I come across something that works for me I will post that info.
Thanks!
Same here. And since Iāve had at least five custom fw builds in the last 7 days alone; fw mattersā¦
Iām getting these same error messagesā¦ What did you do to fix them?
Unfortunately everything I tried, including those from crlogic above, didnāt help and the issue still occurred. Only suggestion I didnāt try was the separate SSID for 2.4ghz as that causes a bunch of headaches for me.
I just finished connecting up some ESP32 devices to the DFRobots and so far things have been better, but itās only been an hour or so since putting them in place.
Donāt overlook firmware. Try the last 10 EAās. No joke.
How does the DFRobot hold up to others? Letās find outā¦
Iām having a tough time getting Arduino SDK to work with my chosen setup of esp32-c3-devkitm-1 in ESPHome. Any chance this can be made to work with the Espressif SDK or is Arduino a requirement?
Seems like a technical possibility. Iāve only just learned enough to make it work w/ ESPHome using Arduino framework. So you would have to translate the Arduino bitsā¦